Vbus is off

We’re experiencing an issue where USB0 will sometimes turn off and stay off until the device is re-imaged. The abridged output from chkusb.sh is:

Linux 002G70315 4.19.94-ti-r42 #1buster SMP PREEMPT Tue Mar 31 19:38:29 UTC 2020 armv7l GNU/Linux
USB is initialized
Warning: CONFIG_USB_OTG defined.
musb-hdrc.0: mode b_idle, Vbus off, timeout 1100 msec
grep: /sys/kernel/debug/musb-hdrc.0: Is a directory
grep: /sys/kernel/debug/musb-hdrc.0.dsps: Is a directory
grep: /sys/kernel/debug/musb-hdrc.1: Is a directory
/sys/kernel/debug/musb-hdrc.0/regdump:Power       : e0
/sys/kernel/debug/musb-hdrc.0/regdump:Testmode    : 00
/sys/kernel/debug/musb-hdrc.0/regdump:DevCtl      : 80
musb-hdrc.1: mode a_wait_bcon, Vbus on, timeout 1100 msec
grep: /sys/kernel/debug/musb-hdrc.0: Is a directory
grep: /sys/kernel/debug/musb-hdrc.0.dsps: Is a directory
grep: /sys/kernel/debug/musb-hdrc.1: Is a directory
/sys/kernel/debug/musb-hdrc.1/regdump:Power       : e0
/sys/kernel/debug/musb-hdrc.1/regdump:Testmode    : 00
/sys/kernel/debug/musb-hdrc.1/regdump:DevCtl      : 19
usb@47401000: peripheral, okay
usb@47401800: host, okay

I’ve read a lot of discussion on what might cause this initially, and we’re looking into possible issues with a rare wandering 5v on the VBus line coming into the beagle, but what I’m not clear on is why this state survives a complete power cycle, and how to reset that state without having to re-image the board.